Leukemia is a leading cause of cancer death and requires a significant expenditure of the national healthcare budget to treat. There is substantial evidence that hereditary factors contribute to leukemia predisposition and the frequently observed familial clustering and racial variations in risk. In many families, affected individuals have developed leukemias of differing type and subtype, suggesting a defect in a gene responsible for hematopoietic stem cell differentiation. A unusual of familial leukemia is that it is inherited with "anticipation" in the form of a declining age of onset with each passing generation, and we hypothesize that a defect in a mitotic clock might be responsible for this phenomenon. A locus for familial AML in association with inherited platelet defects has been mapped to chromosome 21q. Preliminary evidence suggests a second locus for familial AML on chromosome 16q in families that do not have a platelet defect. The specific aims are the following: 1) Clone the locus for familial leukemia on chromosome 21: A. Collect new individuals with the familial platelet disorder/AML syndrome, B. Narrow the critical region on chromosome 21q by studying Down syndrome patients, C. Positionally clone the chromosome 21q locus from leukemia families by mutational analysis of candidate genes from the critical region; 2) Confirm and clone a locus for familial AML on chromosome 16q: A. Use leukemia families to narrow the critical region, B. Use sporadic cases of AML and myelodysplasia to refine the critical region, C. Positionally clone the chromosome 16q AML gene; and c) Characterize the role of familial leukemia genes in hematopoietic stem cell differentiation and leukemia pathogenesis: A. Functionally characterize the genes for familial leukemia, B. Develop animal models, C. Search for common but low penetrance alleles among sporadic leukemia cases.